Immobilization and the risk of venous thromboembolism. A meta-analysis on epidemiological studies.

BACKGROUND: The thrombogenic burden of immobilization remains unknown especially in the medical setting. Most of epidemiological studies estimating the link between risk factors and venous thromboembolism (VTE) have not been designed to evaluate immobilization. The aim of this work was to estimate the risk of VTE in medical bedridden patients by a systematic review and a meta-analysis. METHODS: A research on PUBMED and EMBASE was carried out to retrieve case-control and cohort studies showing the proportion of bedridden patients with or without VTE. Included studies were assigned in six groups according to the following criteria: 1) their design (cohort or case-control), 2) the targeted population (with or without suspicion of VTE) and 3) the medical setting (ambulatory or hospital). Odd-Ratios and Relative Risk for case-control and cohort studies were calculated using a random effect method. Heterogeneity and publication bias were statistically assessed by the I(2) statistics and funnel plots with Egger's tests. RESULTS: 43 studies were included (24181 patients). The pooled RR ranged from 1.46 to 2.77 in the subgroups of cohort studies (n=36) with an overall RR of 1.86 (1.61-2.14; P<0.001). The pooled OR were 2.79 and 2.47 in the two subgroups of case-control studies (n=7), both statistically significant (overall OR: 2.52; 1.70-3.74; P<0.001). Heterogeneity through studies was demonstrated in four subgroups. Publication bias was only observed in one subgroup. CONCLUSIONS: Among medical patients, immobilization increases the risk of VTE. Nevertheless, a specific role of underlying conditions can not be excluded.